[oyi77]

Summary

Proxy Management

• Batch delete: POST /api/settings/proxies/batch-delete — delete multiple proxies in one request
• Auto-test: POST /api/settings/proxies/auto-test — test proxy reachability with optional auto-remove
• Health scheduler: Background scheduler that periodically tests proxies and removes dead ones
• UI improvements: Checkboxes for batch selection, Test All button, colored status indicators

Bug Fix

• resolveProviderAlias: Now follows alias chains transitively (oc → opencode → opencode-zen was broken)
• Removed duplicate opencode alias override

Infrastructure

• Docker compose production config with bind mount
• Proxy health scheduler integrated into server startup

CI Status

• ✅ Lint: PASS
• ✅ Semgrep: PASS
• ✅ i18n Validation: PASS (all locales)
• ✅ Quality Gates: PASS
• ⚠️ Docs Sync: Pre-existing env/docs sync issue (not related to this PR)
• ⚠️ PR Test Policy: Tests not added yet — will add in follow-up commit

Testing

• ✅ Combo auto/all-working routes correctly
• ✅ MiMo models accessible
• ✅ Dashboard functional
• ✅ Docker production build succeeds
• ✅ Batch delete API endpoint exists and requires auth
• ✅ Auto-test API endpoint exists and requires auth

Breaking Changes

None. All changes are additive.

[gemini-code-assist]

Code Review

This pull request integrates the 1ai-ecosystem engineering rules, introduces a background proxy health check scheduler to auto-remove dead proxies, and adds batch delete and auto-test capabilities to the proxy registry. Feedback on these changes identifies several critical issues: a runtime ReferenceError in the auto-test route due to an undeclared alive variable, potential resource leaks in both the auto-test route and the scheduler due to uncleared timeout timers on fetch errors, a portability issue in docker-compose.prod.yml from a hardcoded host path, missing cache invalidation when proxies are auto-removed by the scheduler, and hardcoded English UI strings in the Proxy Registry Manager that violate i18n standards.

Important

The consumer version of Gemini Code Assist on GitHub is being sunset. Starting June 18, 2026, new organization installations will be blocked, and all code review activity will officially cease on July 17, 2026.
For more details on the timeline and next steps, please review the Help Documentation.

[gemini-code-assist]

The variable alive is used on lines 36 and 37 but is never declared or defined in this scope, which will cause a runtime ReferenceError when a proxy test succeeds. Additionally, the timeout timer is not cleared if undiciFetch throws an error, leading to a potential resource leak. Wrapping the cleanup in a finally block and defining alive correctly resolves both issues.

async function testSingleProxy(proxy: { id: string; type: string; host: string; port: number }): Promise<TestResult> {
  const proxyUrl = `${proxy.type}://${proxy.host}:${proxy.port}`;
  const start = Date.now();
  let timeout: ReturnType<typeof setTimeout> | undefined;

  try {
    const controller = new AbortController();
    timeout = setTimeout(() => controller.abort(), TEST_TIMEOUT_MS);
    const dispatcher = createProxyDispatcher(proxyUrl);
    const resp = await undiciFetch(TEST_URL, {
      method: "HEAD",
      signal: controller.signal,
      dispatcher,
      headers: { "User-Agent": "OmniRoute/1.0" },
    });
    const latencyMs = Date.now() - start;
    const alive = resp.status < 500;
    await updateProxy(proxy.id, { status: alive ? "active" : "inactive" }).catch(() => {});
    return { proxyId: proxy.id, host: proxy.host, port: proxy.port, alive, latencyMs };
  } catch (err) {
    const latencyMs = Date.now() - start;
    await updateProxy(proxy.id, { status: "inactive" }).catch(() => {});
    return {
      proxyId: proxy.id,
      host: proxy.host,
      port: proxy.port,
      alive: false,
      latencyMs,
      error: err instanceof Error ? err.message : "Connection failed",
    };
  } finally {
    if (timeout) clearTimeout(timeout);
  }
}

[gemini-code-assist]

Hardcoding a specific host path like /home/openclaw/.omniroute in the production docker-compose file breaks portability and will fail on other environments or for other users. It is highly recommended to use a relative path, an environment variable, or keep the named volume omniroute-prod-data for data persistence.

    volumes:
      - ${OMNIROUTE_DATA_DIR:-./data}:/app/data

[gemini-code-assist]

Import clearDispatcherCache from @omniroute/open-sse/utils/proxyDispatcher so that stale dispatcher instances can be cleared from memory when proxies are auto-removed by the scheduler.

Suggested change

	import { createProxyDispatcher } from "@omniroute/open-sse/utils/proxyDispatcher";
	import { createProxyDispatcher, clearDispatcherCache } from "@omniroute/open-sse/utils/proxyDispatcher";

[gemini-code-assist]

The timeout timer is not cleared if undiciFetch throws an error, leading to a potential resource leak in the event loop. Wrapping the cleanup in a finally block ensures the timer is always cleared.

async function testOneProxy(proxy: { id: string; type: string; host: string; port: number }): Promise<boolean> {
  const proxyUrl = `${proxy.type}://${proxy.host}:${proxy.port}`;
  let timeout: ReturnType<typeof setTimeout> | undefined;
  try {
    const controller = new AbortController();
    timeout = setTimeout(() => controller.abort(), TEST_TIMEOUT_MS);
    const dispatcher = createProxyDispatcher(proxyUrl);
    const resp = await undiciFetch(TEST_URL, {
      method: "HEAD",
      signal: controller.signal,
      dispatcher,
      headers: { "User-Agent": "OmniRoute/1.0" },
    });
    return resp.status < 500;
  } catch {
    return false;
  } finally {
    if (timeout) clearTimeout(timeout);
  }
}

[gemini-code-assist]

When proxies are auto-removed by the scheduler, the proxy dispatcher cache is not cleared. This can leave stale dispatcher instances in memory. It is recommended to call clearDispatcherCache() if any proxies were removed.

  if (removed > 0) {
    try {
      clearDispatcherCache();
    } catch {
      // non-critical
    }
  }

  console.log(
    `${LOG_PREFIX} Sweep complete: ${tested} tested, ${alive} alive, ${removed} auto-removed`
  );

[gemini-code-assist]

The UI labels {selectedIds.size} selected and Delete {selectedIds.size} selected are hardcoded in English. To adhere to the project's internationalization (i18n) standards, these strings should be externalized to the translation files (e.g., en.json) and retrieved using the t() function.

[diegosouzapw]

Thanks @oyi77 — there's real value here (proxy batch-delete, auto-test, health scheduler, alias-resolution fix). Three things before merge: (1) Scope creep — the branch includes a seed 1ai-rules commit adding .cursorrules, AGENTS.md, _rules/* (~1,660 lines), README.md, docker-compose.prod.yml — please drop those; keep only the ~12 proxy feature files. (2) Security gating (hard rule #17) — auto-test makes outbound fetches through user-configured proxies and the scheduler runs periodically, but they live under /api/settings/proxies/, which isn't in LOCAL_ONLY_API_PREFIXES. Please confirm/adjust isLocalOnlyPath() classification (and drop the hardcoded httpbin.org test URL for something configurable/local). (3) Tests — the feature code ships without tests (PR rule #8/#18); please add coverage for batch-delete, auto-test, and the alias fix. A clean re-cut on release/v3.8.44 with those addressed and I'll take it. 🙏